There are two basic technologies used today in color LCDs: active-matrix, commonly called TFT, and passive-matrix, commonly called STN (see definition of these acronyms and other specialized terms used in this article). Comparing the two technologies, TFT is higher contrast (it therefore seems brighter and easier to see), has more saturated colors, is faster (able to display moving images without "ghosting"), can be viewed at wider angles, and is generally more expensive than STN. LCD manufacturers are in general moving away from STN and towards TFT. Sanyo, for example, recently ended all production of STN LCDs and now makes only TFTs.

Any of the three constructions can be combined with either of the two technologies, for a total of six types of color LCDs. Of the six, all except transmissive STN can be used outdoors. Table 2 lists the five remaining LCD types (two of which have variations, for a total of seven types) and shows the number of currently shipping products that use each LCD type, along with the manufacturers of the products.

The number of products that use a given LCD type doesn't necessarily correlate with the quality of the LCD type. Many different factors are involved, including the LCD type's age, cost, power consumption, market demand in a specific product category, ease of integration into a product, system manufacturer's risk tolerance for new technologies, LCD vendor's level of innovation, etc. Table 3 lists the same seven LCD types as Table 2, with outdoor and indoor quality ranking and the primary limitations of each type.

Resolution and size, the remaining two key characteristics, are closely related -- in general, the more pixels (dots) there are in an LCD, the larger it is. Actually the number of dots per inch (dpi) in LCDs shipping in products today ranges from a low of 70 (for a 5.7" display) to a high of 150 (for a 6.7" SVGA). Most current LCDs fall in the range of 90 to 125 dpi. Table 4 lists the 11 standard resolutions that are used in today's portable computers, handhelds, PDAs and phones. For each resolution the table lists the size range (based upon products that are on the market today) and the primary application.

A fifth obscure but important key characteristic of an LCD is aperture ratio. This value, which isn't usually on an LCD spec sheet (but can be calculated from other specs), is the ratio of the area of a pixel (sub-pixel, actually) to its total screen area, which includes all the support structures required by the sub-pixel (e.g., the "transistor" in TFT). A larger aperture ratio means that more light gets through each pixel, which makes the LCD brighter. For a given size LCD, such as 10.4", lower resolutions have a higher aperture ratio because fewer pixels occupy more screen area. This is why you'll often see two different LCD resolutions for indoor and outdoor models of a portable computer. For example, the Panasonic Toughbook 28's standard indoor transmissive TFT is XGA, but the optional indoor/outdoor transflective TFT is only SVGA. Because outdoor light has to go through transflective and reflective LCDs twice, the aperture ratio must be larger to achieve the same relative brightness as a transmissive LCD where the light only has to go through the LCD once. For this reason the LCD vendors tend to offer transflective and reflective LCDs in lower resolutions than transmissive LCDs of the same size. This is also why newer TFT technologies (such as polysilicon, versus the older amorphous silicon) work better as transflective and reflective LCDs -- the support structures in a polysilicon LCD are much smaller, which makes the aperture ratio larger, which in turn makes the LCD brighter. Or, to put it another way, a polysilicon TFT can have higher resolution than an amorphous silicon TFT and still achieve the same brightness level.

Getting down to the meat

Phones

In PC Magazine's review of the Sanyo phone, the screen is described as being "clear inside and even brighter in daylight" -- this is characteristic of a reflective LCD. I exchanged emails with a Nokia 9210 user in Europe, who told me that "to be honest, direct sunlight kills the [9210] display -- it's still useable outdoors, but far from ideal. The screen is so dark in sunlight that wearing any kind of sunglasses makes it almost unreadable." Reflective TFT with a frontlight is therefore the First Choice LCD for phones in Table 5. A white LED light source makes more sense for a phone than a CCFL light source, since a phone doesn't have to be as bright as a handheld computer. Since transflective TFTs aren't available yet in small sizes, transflective STN currently is the best Second Choice. It would have provided a better balance of indoor/outdoor readability in the 9210 than Nokia's modified transmissive TFT -- although it wouldn't be as bright and contrasty indoors.

PDAs

The iPAQ wasn't actually the first PDA to use a reflective TFT. Compaq launched the Aero 2100 PDA in 1999 with a reflective TFT, but the frontlight used white LEDs as the light source (instead of the much brighter CCFL lamp used in the iPAQ). The combination of the very dim light source (10-15 nits) and the first-generation reflective LCD from Sharp caused the product to be rejected by the market. It simply wasn't bright enough either outdoors or indoors.

After reflective TFT, there really isn't a Second Choice LCD. Transmissive STN, such as used in the HP Jornada 520/540 series and the Fujitsu PenCentra 200 CTM, can't be used outdoors at all. Transflective STN just can't compete with reflective TFT on quality terms. Transflective TFT would work in a PDA, but none of the LCD vendors seem to be willing to invest in developing a 3.8" transflective TFT (all transflective TFTs so far are 10.4" and above). Unlit reflective TFT (such as used in Nintendo's very successful GameBoy) won't work in a PDA because PDA users, unlike GameBoy users, aren't willing to go find "task lighting" -- they want to use the PDA wherever they are, whenever they want, without having to go find some light.

Consumer Notebooks

At the end of June, after only two and a half months of shipments, NEC quietly revised the DayLite to use a transflective TFT instead of a reflective TFT. According to an NEC representative, the reason was that "feedback from the target market indicated that users wanted to be able to use the product both outdoors and indoors." Now the DayLite is described by NEC as having "an illuminated reflective LCD, viewable outdoors and indoors." The same NEC representative said that NEC didn't want to use the term "transflective" because consumers didn't easily understand it. Actually, "illuminated reflective" can describe both reflective TFT with a frontlight, or transflective TFT with a backlight, so from the point of view of simplifying things for the consumer, it's a pretty good choice!

After transflective TFT, there really isn't a Second Choice LCD for consumer notebooks. Transmissive TFTs are used in 95% of today's notebooks because the market strongly prefers the high quality of transmissive TFTs (the remaining 5% use transmissive STN). Transflective STNs are even worse than transmissive STNs -- they look so bad compared to transmissive TFTs that nobody would even think about buying one in a notebook today.

Rugged Notebooks

However, now that Panasonic has launched a notebook that uses a 12.1" transflective TFT (the Toughbook 28), the days of modified transmissive TFTs are numbered. They may continue to be used in some situations where the highest available resolution is required (i.e., where the smaller aperture ratio of a high-resolution, amorphous silicon, transflective TFT makes it unacceptably dark). Generally speaking, though, as the range of available sizes of transflective TFTs increase over the next year or two, they will replace most applications of modified transmissive TFTs. An alternative Second Choice LCD for a rugged notebook is still a high-brightness transmissive TFT, particularly if plenty of power is available, such as in a truck or at a fixed outdoor test station on an oilrig.

Vertical Handhelds

In 5"-8.2" vertical handhelds, the situation is very difficult -- the only outdoor color LCD type available in these sizes is transflective STN, which generally has very low market acceptance. (The market acceptance of the reflective STN with frontlight used in one model of Fujitsu's PenCentra is even lower -- a Fujitsu representative told me that it was literally unsaleable until the product was enhanced with a liquid-filled touchpanel, which significantly reduces reflected ambient light. Reflective STN isn't used in any other vertical market handheld in this category, so it really doesn't count.) While none of the existing products in this category have very good outdoor color LCDs, the TouchLite from Two Technologies is probably one of the best. It uses an unusually large 5.7" transflective STN.

LCD choices in this category are limited mostly because LCD vendors can't afford to build LCDs for low-volume markets. Compared to the PDA and consumer notebook markets, the vertical handheld market is miniscule. For an LCD vendor to consider building a new LCD (e.g., a 6" VGA transflective TFT, which many handheld vendors would love to get their hands on), the minimum annual volume commitment is 50K-100K, which is out of the ballpark for vertical handheld manufacturers. In addition, LCD vendors typically want an up-front NRE (non-recurring engineering charge) of one million dollars to cover their development costs, and development takes 9-12 months before the new LCD is in full production. Often LCD vendors won't develop a new LCD, even if the system manufacturer is willing to pay for it, because the business risk or the opportunity cost is too high.

The absence of good outdoor color LCDs in this category may have an interesting side effect on vertical application software. With the proliferation of good 1/4-VGA 3.8" and SVGA 10.4" outdoor color LCDs, software developers may be forced to migrate their applications to one size or the other. Unless Kyocera, the leading developer of transflective STNs, has a real quality breakthrough, transflective STNs may completely die out in this category. The result could be a total lack of any outdoor-color handhelds on which to run vertical software written for half and full-VGA.

Tablet PCs

The 10.4" tablet PC category is in better shape because LCD vendors are more interested in building larger LCDs due to potential applications in industrial, medical and instrumentation markets. Fujitsu recently introduced the Stylistic 3500R, an unlit reflective TFT tablet PC. While this product might seem risky since it can really only be used outdoors, Fujitsu hedges by offering the 3500X (transmissive TFT) and the 3500S (modified transmissive TFT). The 10.4" and 12.1" transflective TFTs just announced by Sharp, Philips and Toshiba should energize this category. I expect to see tablet PC products using these LCDs from several system manufacturers sometime next year.

Conclusions

• LCDs are low margin and thus very price-sensitive products, so very high volume is required

• Many LCD applications outside of consumer notebooks and PDAs are not mobile and therefore can use high-brightness transmissive TFTs -- there's actually not that much demand for outdoor color
• The risks of new product failures in LCDs are high, particularly where the success of a product depends on the user accepting an inherently subjective quality such as sunlight readability
• Market acceptance of new LCD technologies such as transflective TFT are unknown until a system manufacturer has a big hit with a new product, and it's very difficult to get to that point (chicken & egg)
• The cost and time to develop and put into production a new LCD are so enormous, and resources are so limited, that most new products have to be very low risk
• Choosing the sweet spot from a matrix of three constructions x two technologies (with variations such as polysilicon versus amorphous silicon) x 11 resolutions x an average of four sizes per resolution is very difficult -- that's over 500 combinations!

Higher-resolution transflective TFTs (especially the older, amorphous silicon TFTs) are also limited in outdoor brightness by their smaller aperture ratio. Some of the outdoor color LCD types in current use, such as transflective STN, are likely to fade away in the next year or two, which may severely affect vertical handhelds in the 5"-8.2" category, as well as have an unanticipated side effect on vertical application software. Finally, the whole LCD industry moves relatively slowly, so all of these issues are likely to be resolved in a matter of years rather than months. - -- Geoff Walker